Video processing method and apparatus, device and storage medium

Abstract

Embodiments of the present disclosure relate to a video processing method and apparatus, a device and a medium. The method comprises: extracting a video content feature based on an analysis of an original video; obtaining at least one recommended material that matches the video content feature; and generating a target video by processing the original video according to the at least one recommended material, wherein the target video is a video generated by adding the at least one recommended material to the original video.

What is claimed is: 
     
         1 . A video processing method, comprising:
 extracting a video content feature based on an analysis of an original video;   obtaining at least one recommended material that matches the video content feature, comprising:
 determining a playback duration of video frames corresponding to the video content feature in the original video, wherein the video content feature is generated based on a video content of the video frames; 
 tagging the video content feature with a duration identifier based on the playback duration of the video frames; 
 in a case where a plurality of video content features are determined to correspond to the duration identifier, combining the plurality of video content features into a video feature set, and obtaining the at least one recommended material that matches the video feature set, wherein the plurality of video content features comprises the video content feature; and 
 in a case where only the video content feature is determined to correspond to the duration identifier, obtaining the at least one recommended material that matches the video content feature; and 
   generating a target video by processing the original video according to the at least one recommended material, wherein the target video is a video generated by adding the at least one recommended material to the original video.   
     
     
         2 . The video processing method according to  claim 1 , wherein the extracting the video content feature based on the analysis of the original video comprises:
 performing a speech recognition on target audio data of the original video to obtain a text content corresponding to the target audio data; and   performing a semantic parsing on the text content to obtain a first keyword.   
     
     
         3 . The video processing method according to  claim 1 , wherein the extracting the video content feature based on the analysis of the original video comprises:
 performing a sound detection on target audio data of the original video to obtain spectral data corresponding to the target audio data; and   analyzing the spectral data to obtain a second keyword.   
     
     
         4 . The video processing method according to  claim 2 , wherein the target audio data is obtained by:
 obtaining all audio tracks displayed in a video editing application for a video file of the original video;   merging all the audio tracks based on a timeline to generate total audio data; and   comparing a first duration of the total audio data with a second duration of the original video and cropping the first duration of the total audio data to obtain the target audio data in response to the first duration being greater than the second duration, wherein the duration of the target audio data is consistent with the second duration.   
     
     
         5 . The video processing method according to  claim 4 , wherein the merging all the audio tracks based on the timeline to generate the total audio data comprises:
 detecting an audio identifier of each of the audio tracks; and   in response to detecting a target audio track with an identifier representing background music, merging all audio tracks except that target audio track based on the timeline to generate the total audio data.   
     
     
         6 . The video processing method according to  claim 1 , wherein the obtaining the at least one recommended material that matches the video content feature comprises:
 determining a video style feature based on video images of the original video; and   obtaining the at least one recommended material that matches the video style feature and the video content feature.   
     
     
         7 . The video processing method according to  claim 6 , wherein the determining the video style feature based on the video images of the original video comprises:
 performing an image recognition on the video images of the original video, and determining at least one capture object based on a recognition result; and   performing a weighting calculation on the at least one capture object based on preset object weights, and matching a calculation result with preset style classifications to determine the video style feature corresponding to the original video.   
     
     
         8 . The video processing method according to  claim 1 , wherein the generating the target video by processing the original video according to the at least one recommended material comprises:
 setting a material application duration for the at least one recommended material that matches the video content feature according to the duration identifier of the video content feature; and   editing the original video to generate the target video according to the material application duration of the at least one recommended material.   
     
     
         9 . The video processing method according to  claim 8 , wherein the editing the original video to generate the target video according to the material application duration of the at least one recommended material comprises:
 in a case where a material type of the at least one recommended material is a preset target type, obtaining target video frames corresponding to the material application duration of the at least one recommended material from the original video;   performing an image recognition on the target video frames to obtain a main region of a capture object;   determining a material region for adding the at least one recommended material on the target video frames based on the main region of the capture object; and   editing the original video to generate the target video according to the material application duration and the material region of the at least one recommended material.   
     
     
         10 . The video processing method according to  claim 5 , wherein the detecting the audio identifier of the each of the audio tracks comprises:
 identifying a sound feature of an audio corresponding to the each of the audio tracks;   matching the sound feature of the audio corresponding to the each of the audio tracks with a sound feature corresponding to each preset audio identifier; and   determining the audio identifier of the each of the audio tracks based on a matching result.   
     
     
         11 . The video processing method according to  claim 1 , wherein the obtaining the at least one recommended material that matches the video feature set comprises:
 inquiring a preset correspondence to determine whether there is an enhancement material corresponding to the video feature set according to the video feature set;   in case that there is no enhancement material corresponding to the video feature set, splitting the video feature set into individual content features to match the at least one recommended material; and   in case that there is an enhancement material corresponding to the video feature set, taking the enhancement material as the at least one recommended material corresponding to the video feature set.   
     
     
         12 . An electronic device, comprising:
 a processor;   a memory for storing executable instructions by the processor,   wherein the processor is configured to read the executable instructions from the memory and execute the executable instructions to:   extract a video content feature based on an analysis of an original video;   obtain at least one recommended material that matches the video content feature, comprising:
 determining a playback duration of video frames corresponding to the video content feature in the original video, wherein the video content feature is generated based on a video content of the video frames; 
 tagging the video content feature with a duration identifier based on the playback duration of the video frames; 
 in a case where a plurality of video content features are determined to correspond to the duration identifier, combining the plurality of video content features into a video feature set, and obtaining the at least one recommended material that matches the video feature set, wherein the plurality of video content features comprises the video content feature; and 
 in a case where only the video content feature is determined to correspond to the duration identifier, obtaining the at least one recommended material that matches the video content feature; and 
   generate a target video by processing the original video according to the at least one recommended material, wherein the target video is a video generated by adding the recommended material to the original video.   
     
     
         13 . The electronic device according to  claim 12 , wherein the extracting the video content feature based on the analysis of the original video comprises:
 performing a speech recognition on target audio data of the original video to obtain a text content corresponding to the target audio data; and   performing a semantic parsing on the text content to obtain a first keyword.   
     
     
         14 . The electronic device according to  claim 12 , wherein the extracting the video content feature based on the analysis of the original video comprises:
 performing a sound detection on target audio data of the original video to obtain spectral data corresponding to the target audio data; and   analyzing the spectral data to obtain a second keyword.   
     
     
         15 . The electronic device according to  claim 13 , wherein the target audio data is obtained by:
 obtaining all audio tracks displayed in a video editing application for a video file of the original video;   merging all the audio tracks based on a timeline to generate total audio data; and   comparing a first duration of the total audio data with a second duration of the original video and cropping the first duration of the total audio data to obtain the target audio data in response to the first duration being greater than the second duration, wherein the duration of the target audio data is consistent with the second duration.   
     
     
         16 . The electronic device according to  claim 15 , wherein the merging all the audio tracks based on the timeline to generate the total audio data comprises:
 detecting an audio identifier of each of the audio tracks; and   in response to detecting a target audio track with an identifier representing background music, merging all audio tracks except that target audio track based on the timeline to generate the total audio data.   
     
     
         17 . The electronic device according to  claim 12 , wherein the obtaining the at least one recommended material that matches the video content feature comprises:
 determining a video style feature based on video images of the original video; and   obtaining at least one recommended material that matches the video style feature and the video content feature.   
     
     
         18 . The electronic device according to  claim 17 , wherein the determining the video style feature based on the video images of the original video comprises:
 performing an image recognition on the video images of the original video, and determining at least one capture object based on a recognition result; and   performing a weighting calculation on the at least one capture object based on preset object weights, and matching a calculation result with preset style classifications to determine the video style feature corresponding to the original video.   
     
     
         19 . A non-transitory computer-readable storage medium stored thereon a computer program, which when executed by a processor, causes the processor to:
 extract a video content feature based on an analysis of an original video;   obtain at least one recommended material that matches the video content feature, comprising:
 determining a playback duration of video frames corresponding to the video content feature in the original video, wherein the video content feature is generated based on a video content of the video frames; 
 tagging the video content feature with a duration identifier based on the playback duration of the video frames; 
 in a case where a plurality of video content features are determined to correspond to the duration identifier, combining the plurality of video content features into a video feature set, and obtaining the at least one recommended material that matches the video feature set, wherein the plurality of video content features comprises the video content feature; and 
 in a case where only the video content feature is determined to correspond to the duration identifier, obtaining the at least one recommended material that matches the video content feature; and 
   generate a target video by processing the original video according to the at least one recommended material, wherein the target video is a video generated by adding the recommended material to the original video.   
     
     
         20 . The non-transitory computer-readable storage medium according to  claim 19 , wherein the extracting the video content feature based on the analysis of the original video comprises:
 performing a speech recognition on target audio data of the original video to obtain a text content corresponding to the target audio data; and   performing a semantic parsing on the text content to obtain a first keyword.